Nasogastric hydration versus intravenous hydration for infants with bronchiolitis: Which route is best?
Nasogastric hydration versus intravenous hydration for infants with bronchiolitis: a randomised trial
The Lancet (Respiratory Medicine), Early Online Publication, 21 December 2012,
Ed Oakley, Meredith Borland, Jocelyn Neutze, Jason Acworth, David Krieser, Stuart Dalziel, Andrew Davidson, Susan Donath, Kim Jachno, Mike South MD, Theane Theophilos, Franz E Babl, for the Paediatric Research in Emergency Departments International Collaborative (PREDICT)
The article is available here.
An expert comment on the paper is here.
The main care for infants with moderate or severe bronchiolitis is supportive, with maintenance of hydration for those with inadequate feeding (about a third of admitted patients)
and administration of supplemental oxygen and other respiratory support (eg CPAP, IPPV) as required.
For a long time the appropriate route of administration of fluid to maintain hydration has been hotly debated, with often strongly-held views over the IV and nasogastric routes.
This multi-centred Australian RCT is the first study to directly address this question. It enrolled 759 subjects aged 2 to 12 months
The study demonstrated no difference in the primary endpoint (length of hospital stay), nor in: time until medically ready for discharge, admission to the intensive-care unit, need for ventilation or continuous positive airway pressure support, oxygen desaturation, bradycardia, and apnoea. No patient had pulmonary aspiration in either group.
The main differences demonstrated were in the practicalities of insertion of an NG tube vs an IV cannula.
Success rate on the 1st attempt was 85% for NG tube vs 56% for an IV line. Higher proportions in the IV group needed 2 and 3 or more attempts at insertion.
The study demonstrates that either route of fluid administration appears effective and safe but that access for NG fluid administration is easier to achieve. Fewer attempts at insertion make the procedure less uncomfortable for the infant and less stressful for clinical staff. Infant distress during handling for insertion of an access device often leads to additional respiratory effort in an infant who is already working hard to breathe. It makes sense that fewer attempts is better. In many settings, including resource poor areas, NG tubes can be inserted by nursing staff which may provide less expensive and more timely hydration treatment.
Moving from evidence to anecdote: The study didn’t address one of my pet issues – that babies with bronchiolitis who are fed via NG tube appear to cry (from hunger) less than those who are starved and given IV fluids. This didn’t appear translate into any measured outcomes in this study but I still think it is matter of interest to a hungry unwell baby! I can recall numerous occasions when an infant with bronchiolitis has been admitted to our ICU for possible respiratory support but following a period of minimal handling and some nasogastric feeding has settled to sleep and additional respiratory therapy has not been required.
It is good to see research into these basic everyday clinical care questions.
Bronchiolitis is the most common lower respiratory tract infection in infants and the leading cause of hospital admission. Hydration is a mainstay of treatment, but insufficient evidence exists to guide clinical practice. We aimed to assess whether intravenous hydration or nasogastric hydration is better for treatment of infants.
In this multicentre, open, randomised trial, we enrolled infants aged 2—12 months admitted to hospitals in Australia and New Zealand with a clinical diagnosis of bronchiolitis during three bronchiolitis seasons (April 1—Oct 31, in 2009, 2010, and 2011). We randomly allocated infants to nasogastric hydration or intravenous hydration by use of a computer-generated sequence and opaque sealed envelopes, with three randomly assigned block sizes and stratified by hospital site and age group (2—<6 months vs 6—12 months). The primary outcome was length of hospital stay, assessed in all randomly assigned infants. Secondary outcomes included rates of intensive-care unit admission, adverse events, and success of insertion. This trial is registered with the Australian and New Zealand clinical trials registry, ACTRN12605000033640.
Mean length of stay for 381 infants assigned nasogastric hydration was 86·6 h (SD 58·9) compared with 82·2 h (58·8) for 378 infants assigned intravenous hydration (absolute difference 4·5 h [95% CI −3·9 to 12·9]; p=0·30). Rates of admission to intensive-care units, need for ventilatory support, and adverse events did not differ between groups. At randomisation, seven infants assigned nasogastric hydration were switched to intravenous hydration and 56 infants assigned intravenous hydration were switched to nasogastric hydration because the study-assigned method was unable to be inserted. For those infants who had data available for successful insertion, 275 (85%) of 323 infants in the nasogastric hydration group and 165 (56%) of 294 infants in the intravenous hydration group required only one attempt for successful insertion.
Intravenous hydration and nasogastric hydration are appropriate means to hydrate infants with bronchiolitis. Nasogastric insertion might require fewer attempts and have a higher success rate of insertion than intravenous hydration.